Ed Garguilo's orginal user accuracy recipe

From: Elaine Hunt (ehunt@ces.clemson.edu)
Date: Tue Feb 23 1999 - 23:30:11 EET

This is Ed's paper as it was submitted to the user group for consideration.
Some of it may seem a bit strange to new users.

E. P. Gargiulo - Du Pont Company - June 1990
Studies of user build strategies using a common test article permit the entire
user community to share experience and compare results against a de facto
standard. I have initiated a study of the planar dimensional accuracy of
StereoLithography. This study is aimed at finding the effects of build
strategy (cross hatch and layer thickness choices), beam compensation, scanner
accuracy, polymers, and cure techniques on XY writing accuracy. Only in-plane
dimensional error will be addressed in this study because out-of-plane warping
introduces too many variables at this point in our understanding. In-plane
errors caused by the lack of beam compensation, polymer shrinkage and machine
variations will lay a good groundwork for future studies. You can participate
in this study and help us to get a user accuracy baseline.

At the May users' meeting I displayed a 9 inch square, plate-like article for
discussion. This geometry has the advantage of using the entire bath area and
giving a large number of dimensional features, but takes more than 2.5 hrs to
slice (NEC 386) and 15 hrs to build (0.100 in. X-60-120 hatch, 0.01 in.
layers). It is also subject to considerable warping. Several of us reworked
the design of this article to reduce its tendency to warp and hopefully
make it
less time consuming to build. I have built the revised part several times and
it does seem more warp resistant.

The .STL file for the new model is available to all users. I would like to
have each user build the model to the same parameters set out below so we can
establish a user accuracy baseline. These parts will be influenced by machine
differences, polymer types and conditions, environmental conditions
(temperature and humidity), and cleaning and cure processes that are largely
uncontrollable. If you send me the part (carefully packed to prevent
breakage), I will measure it on a coordinate measuring machine and include it
in a summary report.
The following standard build conditions are requested:

SLICE: (File USER.UII and USER_S.UII on diskette)
        Support Part
          Output Scale 1.0 1.0
Resolution 5000 5000
Layer Thickness 0.010 0.010 in. Fixed
 X Hatch Spacing 0.25 0.075
Y Hatch Spacing 0.25 0.000 in.
60/120 Hatch Spacing 0.00 0.075 in.
X Skin Fill Spacing 0.00 0.004 in.
Y Skin Fill Spacing 0.00 0.000 in.
        MSA 0 50
                MIA 0 0
(Slice time required on NEC approx. 100 mm.)

PREPARE: (File USER.R on diskette)
PARAMETERS MANAGER: (Take all defaults)
        XY Shrink 0.000
        Z Shrink 0.000
        Number of Copies 1
        Multi-part Spacing 0.000
No Build Options Selected

RANGE MANAGER: (See Prepare Report for 19.5 mw laser power)
Add Z Range at 2400 slice units.
Add z Range at 3750 slice units.
SP Calculations:(Material File XB5081.NAT on diskette) Boundary and Hatch
Thickness 0.016 in. Skin Fills 0.022 in.
Range 1: NS 0;ZW 30
Range 2: NS 1;ZW 120;P1 10
Range 3: NS 1;z 60;P1 10
Range Defaults ZD 8;z 0.2;ZA 0.2

BUILD: (All defaults) (USER.PRN and xB5081.NAT on diskette)
Our bath temperature is 33C.

(Build time is approx. 16 hrs. 45 mm.)

CLEAN: Wash in isopropyl alcohol.
Ultrasonic isopropyl alcohol bath for 5 mm.
Drain dry.
Quick rinse in methanol.

CURE: Six minutes in air as follows:
Part on Elevator Stage
90 sec side one (supports down)
90 sec side two (supports up)

Remove from Elevator Stage and wash in methanol to remove debris from support
removal. Blot off remaining fluid.
90 sec side one
90 sec side two

NOTE: I have observed some delamination on the back edge of the part at the
base approximately 6 layers into the part. This delamination should have no
effect on the in-plane measurements.
Not everyone will be able to reproduce our cleaning/curing process. Please
detail, on the attached data sheet, your cleaning and curing process.

If you would like to submit a second part made to your most accurate recipe, I
will also measure it and report results at our next meeting.

Opinions, suggestions, and other controversial matter VOID where prohibited.
Elaine T. Hunt, Director
Clemson University Laboratory to Advance Industrial Prototyping
206 Fluor Daniel Bldg. Clemson, SC 29643-0925
864-656-0321 (voice) 864-656-4435 (fax)

For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:51:03 EEST